Neural network localization in Parkinson's disease with impulse control disorders

Abstract

Background: There is a huge heterogeneity of magnetic resonance imaging findings in Parkinson's disease (PD) with impulse control disorders (ICDs) studies. Here, we hypothesized that brain regions identified by structural and functional imaging studies of PD with ICDs could be reconciled in a common network.

Methods: In this study, an initial systematic literature review was conducted to collect and evaluate whole-brain functional and structural magnetic resonance imaging studies related to PD with ICDs. We subsequently utilized the Human Connectome Project (HCP) dataset (n = 1,093) and a novel functional connectivity network mapping (FCNM) technique to identify a common brain network affected in PD with ICDs.

Results: A total of 19 studies with 25 contrasts, incorporating 345 individuals with PD and ICDs, and 787 individuals with PD without ICDs were included in the analysis. By using the HCP dataset and a novel FCNM technique, we ultimately identified that the aberrant neural networks predominantly involve the default mode network (middle and inferior temporal gyrus, anterior cingulate cortex, angular gyrus) and subcortical network (caudate nucleus).

Conclusion: This study suggests that the heterogeneous neuroimaging findings in PD with ICDs can be attributed to shared abnormalities in the default mode and subcortical networks. These dysfunctions are associated with impaired self-regulation, decision-making, and heightened impulsivity in PD with ICDs. Our findings integrate diverse neuroimaging results from previous studies, providing a clearer understanding of the neurobiological mechanisms underlying PD with ICDs at a network level.

Keywords: Parkinson’s disease; functional connectivity network mapping; impulse control disorders; network localization; voxel-based morphometry.

Figure 1

PD with ICDs brain dysfunctional networks based on 4-mm radius sphere dysfunctional networks are shown as network probability maps thresholded at 60%, showing brain regions functionally connected to more than 60% of the contrast seeds. PD with ICDs, Parkinson’s disease with impulse control disorders.

Figure 2

Associations of dysfunctional brain networks with canonical brain networks in PD with ICDs based on 4-mm radius sphere. Polar plots display the proportion of overlapping voxels between each brain dysfunctional network and a canonical network relative to all voxels within the corresponding canonical network. The red circles represent brain dysfunction networks, defined as significant networks, exhibiting ≥10% overlap with canonical networks, whereas the blue circles represent non-significant networks with <10% overlap. PD with ICDs, Parkinson’s disease with impulse control disorders.